The present invention relates to a streak apparatus suitable for use in such applications as the measurement of time-dependent distribution of intensity in phenomena such as light emission.
A streak apparatus employs a streak tube for measuring the time-based intensity distribution of incident light by converting it into a spatial distribution of intensity on a phosphor screen. Since the streak apparatus can provide a time resolution of up to the order of picoseconds, it is particularly adaptive to analysis of light emission whose intensity changes at extremely short intervals.
The composition and operation of a conventional streak tube are first described hereinafter.
FIG. 1A is a cross sectional view of a conventional streak tube as viewed from the side of the photocathode, and FIG. 1B is a longitudinal sectional view of the streak tube taken on a plane that includes the tube axis and is parallel to the deflecting electrodes. FIG. 2A is another cross sectional view of the conventional streak tube as viewed from the side of the photocathode, and FIG. 2B is a longitudinal sectional view of the streak tube taken on a plane that includes the tube axis and is perpendicular to the deflecting electrodes. FIG. 3 is a diagram showing the profile of a voltage to be applied to the deflecting electrodes. In these figures, reference numeral 1 is a vacuum air-tight container, 2 is an incident window, 3 is a photocathode, 4 is a mesh electrode, 5 is a focusing electrode, 6 is an anode plate (aperture electrode), 8 is an exit window, 9 is a phosphor screen, 10 is an optical image; 11 is the principle orbit of electron beam, and 12 is a .beta.-orbit of an electron beam which is defined hereinafter.
As shown in the figures, the vacuum air-tight container 1 of the streak tube is provided at one end thereof with an incident window 2 upon which an optical image to be analyzed is incident, and at the other end thereof with an exit window 8 from which the processed optical image emerges. The incident window 2 is provided with a photocathode 3 on the inner surface thereof, and the exit window 8 is provided with a phosphor screen 9 on the inner surface thereof. A mesh electrode 4, a focusing electrode 5, anode 6 and deflecting electrodes 7 are arranged in this order between the photocathode 3 and the phosphor screen 9 along the tube axis of the container 1.
The photocathode 3, focusing electrode 5, mesh electrode 4 and the anode 6 having a central aperture are supplied with increasing voltages in this order and the phosphor screen 9 and the anode 6 are connected to the same potential. Assuming that a linear optical image 10 is projected onto the photocathode 3 through the incident window 2 by means of a certain device (not shown) in such a manner that the image passes through the center line of the photocathode 3. The photocathode 3 then emits a photoelectron image that corresponds to the optical image. The emitted electrons are accelerated with the mesh electrode 4, are focused with the focusing electrode 5, pass through the central aperture in the anode 6, and travel through the gap space between the deflecting electrodes 7 toward the phosphor screen 9.
The deflecting electrodes 7 between which the linear electron image passes are supplied with, for example, a deflection voltage that changes with time as shown in FIG. 3. The direction of the electric field produced by this voltage is perpendicular to the tube axis and the linear electron image (i.e., in the direction perpendicular to the surface of the drawing of FIG. 1B) and its intensity is proportional to the applied deflection voltage. The deflection voltage allows the linear electron beam to be swept across the phosphor screen 9 in a direction perpendicular to the line of the linear electron beam (i.e., in the direction indicated by arrow S in FIG. 2B) in such a manner that linear output light images corresponding to the linear optical image projected on to the photocathode 3 are successively arranged on a time basis in a direction perpendicular to the longitudinal direction of the line, thereby producing a "streak image" on the phosphor screen 9. Therefore, by measuring quantitatively the change in luminance in the direction in which the streak image is formed, that is, in the sweep direction, the time-dependent change in the intensity of the optical image incident on the photocathode 3 is determined. As the width (W) of the linear output light image obtained on the phosphor screen is narrowed, the time resolution that can be attained is higher.
Many photoelectrons are emitted from a point on the photocathode 3 at the same time at various angles with various energies. For instance, photoelectrons emitted from point P of the linear optical image on the photocathode 3 have an initial energy distribution ranging from zero to several electron volts. The direction of their emission also ranges from the one that is perpendicular to the photocathode 3 to those having angles of .theta.(0.degree. &lt;.theta.&lt;90.degree. ) with respect to the normal to the photocathode 3. If these photoelectrons are allowed to travel straightforwardly, they will become divergent to produce a blurred electron image. Therefore, in order to focus the divergent photoelectrons on the phosphor screen, an electron lens formed by a focusing electrode is employed.
The following description is made on transitorbits of photoelectrons when no deflection voltage is applied between a pair of deflecting plates 7a and 7b. If the photoelectrons are emitted from the photocathode 3 with an initial energy of zero, they will transit on the orbit 11 which is hereunder referred to as the principle orbit. The other electrons will travel on the orbits 12 which are hereunder referred to as .beta.-orbits. There are an infinite number of .beta.-orbits 12 and they depart farther away from the principle orbit 11 as photoelectrons are emitted from the photocathode at larger angles of .theta. with respect to the normal to the photocathode and as they have greater initial energies. For the sake of convenience of the following description, the .beta.-orbits 12 are to be represented by those of the photoelectrons with an energy of 1 eV that are emitted from the photocathode 3 symmetrically with the principle orbit 11 at an angle of 60.degree. with respect to the normal to the photocathode 3.
If it is supposed that the photoelectrons traveling on the principle orbit 11 impinge on the phosphor screen 9 at point P', the photoelectrons traveling on any .beta.-orbit 12 can be allowed to impinge on the phosphor screen at a point in substantial coincidence with point P' by properly adjusting the voltage to be applied to the focusing electrode 5.
As shown generally in FIG. 2B, photoelectrons traveling on the .beta.-orbit 12 first depart away from the principle orbit 11 but are thereafter subjected to the force acting toward the principle orbit that is produced by the electron lens formed of the focusing electrode 5. Therefore, the distance between the principle orbit 11 and the photoelectrons traveling on the .beta.-orbits 12 becomes the greatest in the vicinity of the end of the focusing electrode 5 which is on the side of the aperture electrode 6. The photoelectrons are then subjected to the velocity component acting toward the principle orbit 11 and the focusing effect of the electron lens disappears in the middle area between the aperture electrode 6 and the deflecting electrodes 7. Thereafter, the photoelectron beam makes a linear motion at a velocity substantially corresponding to the voltage applied between the photocathode 3 and the aperture electrode 6 and approaches the principle orbit until it impinges on the phosphor screen 9 at point P'. In this way, the photoelectrons that were emitted from the photocathode at various angles and with various energies are focused at a point that substantially coincides with point P', thereby decreasing the width of beam spread, W, on the phosphor screen.
However, if a sweep voltage is applied between the deflecting electrodes, the width of beam spread, W, on the phosphor screen will increase in the sweep direction. Let us first discuss this problem with reference to the static case where the change in deflection voltage is so slow that it may be regarded as zero during the passage of an electron beam through the deflection space.
FIG. 4 shows only the deflecting electrodes and the phosphor screen as isolated from the other components of the streak tube. Beam A in FIG. 4 denotes photoelectrons travelling on the principle and .beta.-orbits when a deflection voltage of +VD (positive) is applied to the deflecting plate 7a and a voltage of -VD (negative) applied to the deflecting plate 7b. Beam C denotes photoelectrons traveling on the principle and .beta.-orbits when a negative voltage of -VD is applied to the deflecting plate 7a and a positive voltage of +VD applied to the deflecting plate 7b. Each of the beams A and C is the result of deflection of beam B which would be focused on a single point on the phosphor screen 9 in the absence of deflection voltage. As shown, each of the beams A and C produces a spread on the phosphor screen 9 which results from the difference between the amounts of deflection of electrons (a) and (b) on .beta.-orbits.
FIG. 5 shows equipotential surfaces created around the deflecting electrodes 7 when the absolute value of VD is 500 volts. As for the electron beam A, electron (a) traveling on .beta.-orbit near the input end of the deflecting electrodes 7 is closer to the deflecting plate 7a supplied with +500 volts, so that this electron is accelerated in the direction of the tube axis by the force Fa. On the other hand, electron (b) traveling on another .beta.-orbit is closer to the deflecting plate 7b supplied with -500 volts, so that this electron is decelerated by the force Fb. As a result, electron (b) will travel through the deflection field more slowly than electron (a) and hence is deflected by a greater amount than electron (a). This relationship between electrons (a) and (b) is reversed in the case of beam C. As a consequence, electron beams traveling toward the peripheral edge of the phosphor screen 9 will be focused in the front of the phosphor screen 9 and produce a spread Ws when they encounter the screen. The amount of this beam spread increases as the beam is deflected farther away from the center of the phosphor screen 9. The beam spread Ws is one of the major causes of deteriorated time resolution of the streak tube.
Assuming the sweeping speed of the beam on the phosphor screen 9 is represented by Vs, the time resolution of streak tube, .DELTA.t, as defined by the beam spread Ws is expressed by: EQU .DELTA.t=Ws/Vs.
If a beam spread occurs only in the area having a large deflecting angle, there will be no problem so long as the effective central area of the phosphor screen and the range of small deflection angles are selectively used.
Let us then consider the dynamic case where the change in deflection voltage is so rapid that it cannot be substantially neglected when electrons are traveling through the deflection space.
FIG. 6 shows the waveforms of deflection voltages to be applied to the deflecting electrodes. When the deflecting plate 7a is supplied with a linearly changing ramp voltage Vd.sub.1 (t) and the deflecting plate 7b is supplied with another ramp voltage Vd.sub.2 (t), the voltage applied between the two deflecting plates at time t is expressed by: Vd.sub.1 (t)-Vd.sub.2 (t).
If the change in two ramp voltages that occurs when photoelectrons are traveling between the deflecting electrodes is negligibly smaller than the beam accelerating voltage applied between the photocathode and the aperture electrode, the conclusion already stated in connection with the application of a dc deflection voltage can be applied to this case.
If photoelectrons traveling from the photocathode to the aperture electrode are accelerated to about 10 keV, the velocity component parallel to the tube axis is about 6.times.10.sup.7 m/s in the area of the deflecting electrodes. If the deflecting electrodes are 12 mm in length, it will take about 200 ps for the photoelectrons to pass through the gap space between the two deflecting electrodes. Therefore, if the change in the ramp voltage shown in FIG. 6 is no more than about 3 kV per .mu.s, the change in the deflection voltage that occurs when photoelectrons are passing through the gap space between the deflecting electrodes is only about 0.3 V, which is so much smaller than the accelerating voltage of 10 keV that one may safely disregard it and apply the theory applicable to the case where a dc deflecting voltage is supplied. However, if the deflection voltage changes by as much as 3 kV during the period of 200 ps when photoelectrons are passing through the deflection field, the beam spread on the phosphor screen will come out in a somewhat different manner.
FIG. 7 shows how an electron beam spreads when a substantial change occurs in the deflection voltage. As in the case of the application of a dc deflection voltage, electrons moving slowly in the direction of tube axis are more subject to the action of the deflection field than fast moving electrons. However, when the deflection voltage is changing very rapidly, electrons (a) and (b) (on .beta.-orbits) in each of the beams A, B and C have a different velocity relationship in the direction of tube axis than in the case where a dc deflection voltage is applied, and a very rapidly changing ramp voltage is also exerted upon beam B traveling toward the center of the phosphor screen 9.
Referring to the case as shown in FIG. 7, beam B passing between the deflecting electrodes is initially deflected toward deflecting plate 7a since at the time of beam incidence, a positive voltage is applied to the deflecting plate 7a and a negative voltage applied to the deflecting plate 7b. In the mean time, the potential relationship between the plates 7a and 7b is reversed and the beam B is deflected toward the plate 7b. As a consequence of these changes in deflection voltage, the beam B is finally converged to reach the center of the phosphor screen.
Such a rapidly and greatly changing deflection voltage will cause different effects on the orbits of electrons than when a dc deflection voltage is applied and it is not easily determined what the beam spread will become unless electron orbits are analyzed by computer simulation. According to analysis by this technique, the beam spread Wd is the widest on the center of the phosphor screen and the way in which an electron beam spreads on the phosphor screen also differs more considerably at large deflection angles than that in the case where a dc deflection voltage is applied.
Computer simulation has also shown that the beam spread that occurs when a rapidly changing deflection voltage is applied is substantially the same as what is produced on the phosphor screen when a dc deflecting voltage is applied and the voltage being applied to the focusing electrode is adjusted in such a manner that a photoelectron beam will be focused on a plane ahead of the phosphor screen as indicated by dashed lines in FIG. 7. Therefore, if a preadjustment is made in such a manner that in a static state, an electron beam will be focused at a point (static focal point F in FIG. 8) which lies on the center line passing through the phosphor screen, but is suitably in the front of the phosphor screen the beam can accurately be focused on the phosphor screen in sweep mode, thereby preventing the beam spread that would be created by the deflection voltage. This technique has been conventionally used to prevent the beam spread that occurs when a rapidly changing deflection voltage is applied.
The focusing plane can be adjusted by changing the voltage to be applied to the focusing electrode; if the voltage on the focusing electrode is shifted in the negative direction, the focusing plane is shifted to a position ahead of the phosphor screen (i.e., toward the photocathode). In this way, the position of a focusing plane can be adjusted by adjusting the voltage on the focusing electrode.
However, the degree by which the beam focusing point is shifted ahead of the phosphor screen as a result of sweeping operation, differs with the sweep speed and, therefore, the amount by which the voltage applied on the focusing electrode must be shifted in the negative direction is also dependent on the sweep speed. In other words, a change in sweep speed necessitates a corresponding adjustment in the voltage applied to the focusing electrode. As a further problem, if a light beam to be measured contains extremely intense light pulses, a photoelectron beam of high density is produced and the repulsion of emitted electrons by the space-charge effect causes the diameter of the beam to be increased in sweep direction (i.e., the distance between electrons (a) and (b) in FIGS. 4 and 5) when it is passed between the deflecting electrodes. Unlike the increase in beam size that occurs on account of the energy or angle of electron emission from the photocathode, the beam size that has increased as a result of repulsion of photoelectrons during its transit period from the photocathode to the deflection electrodes has to be compensated by applying another focusing voltage. Therefore, in practice, it is very difficult to adjust the voltage on the focusing electrode in accordance with the change in the intensity of a light beam to be measured.
Another problem with the increase in the intensity of light pulses to be measured is that it causes a spread of the photoelectron beam not only in the sweep direction but also in the direction of the axis of the streak tube, and this is another cause of deterioration in the time resolution that can be attained by the streak tube. FIG. 9 illustrates how a photoelectron beam spreads in the direction of the tube axis when an extremely intense light impulse is applied to the streak tube. Upon incidence of such a light impulse, pulses of a very narrow width in the form of clouds of a great number of photoelectrons are emitted from the photocathode in the direction of tube axis, and by the time when they are launched between the deflecting electrodes, they will have spread in the direction of the tube axis by the space-charge effect as shown by a line A-B. If such spread pulses are deflected by a ramp voltage applied to the deflecting electrodes, a pulse spread will be produced on the phosphor screen as indicated by line A'-B'.